www.nettime.org
Nettime mailing list archives

<nettime> dox vs hax digest [phillips, geer]
nettime's_hackumentarist on Tue, 2 Mar 2004 03:49:02 +0100 (CET)


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

<nettime> dox vs hax digest [phillips, geer]


Re: <nettime> floss enforcement/compliance
     ed phillips <ed {AT} cronos.net>
     Benjamin Geer <ben {AT} socialtools.net>

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Date: Mon, 1 Mar 2004 13:45:26 -0800
From: ed phillips <ed {AT} cronos.net>
Subject: Re: <nettime> floss enforcement/compliance

Thanks again, Ben,

But you are wrong here as far as the GPL is concerned and in my
opinion as to the benefits of releasing a trivial hack. First of all,
releasing a trivial hack is in no way compelled by the GPL.
What may seem to be a very minor part of the GPL is in fact a big part
of the reason why free software is so widely used. No one is compelled
to release the specific uses they make of free software tools, unless
they distribute the specific use to the general public. Therefore,
your example is wrong in spirit and letter.

Yes, I've heard these IBM fellows on numerous panels give the same
kind of laughter producing schtick. But, what trivial hacks have you
actually seen IBM release to the networks? On every single one of the
free software projects that they have contributed to, they have spent
considerable amounts of money and time documenting and peer reviewing
their code before they release it. I would expect no less of them.

To the government of Foobaruda, I would say, It is not necessary for
you to release the specific uses that you make of free software to the
public for whatever reason you have, whether it be because the code is
not good enough or not generic enough to be of use by others, or even
if you have some fantasy that there is something specific to what you
are doing with this free software that you don't want the world to
see. I would say to them, Use some of them money you have saved to pay
someone to substantively contribute to a free software project, either
through documentation or
through the contribution of generic modifications, bugfixes, or
toolsets. The key concept here is generic. If you don't yet have
anything generic to contribute, get on a mailing list and offer some
help. Edit some of the existing documentation. Post a useful idiom.
But please don't just release code you were too lazy or too harried to
review or to make generic enough to be of use to anyone else.

As someone who is out here promoting and using free software in the actually
existing networks in more than a merely academic way, this aspect of
the GPL has been one of the most important. The GPL has allowed me to
tell my clients, whether governmental or private, that their specific
code is theirs in the sense that they are not compelled to release it
to the general public but that none of the generic problems that I solve
belong to them or can be considered their intellectual property, and
that in fact generic tools and solutions should be contributed back to
the community of users from which they have benefitted.

We would never have convinced the large number of private and
governmental users of free software that it made sense to them to use
it if they were compelled to release any and all software that they
develop for their own use. Nor would a flood of released rushed and
half-baked hacks be very helpful to anyone.

An important distinction needs to be made between a rush job hack and
a well documented, generic tool. A rush-job hack gets the job done and
in compiled form, gets the machine to do exactly what it needs to do.
A well documented, generic tool is friendly to human consumption an in
addition to getting the job done, it's source is readable and well
designed. rush job hacks deserve to only be read by brave deep code
divers willing to rewrite them; they need not be inflicted on the public.

I'm not saying that a law should be passed disallowing the release of
said trivial hacks. You are free to do as you please.

Nonetheless, I think that users of free software need to begin to
think differently about what it is they are developing and what of 
substance they can contribute back. One good document is worth a
thousand trivial hacks. A generic tool is worth a thousand specific
uses of it....
 
On Sun, Feb 29, 2004 at 11:36:05AM +0000, Benjamin Geer wrote:

> ed phillips wrote:
> > If you are the
> > government of Extremadura and you need to release an application to
> > all your far flung Linux servers that
> > does foo and stores bar in a MySQL database, you are free to
> > distribute this on all your computers without formally having to
> > inflict the ugly, just good enough to get the job done hack on the
> > rest of us. Nor do you have to pay MySQL 450 or whatever dollars per
> > install of the application.
> 
> Easy solution: release the source code with a big notice on it: 'This 
> code is crap; we're just releasing it to comply with the GPL.' :)  If 
> they're not comfortable admitting that their code is crap, perhaps the 
> GPL will act as an incentive to write better code in the first place.
 <...>

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Date: Tue, 02 Mar 2004 00:23:33 +0000
From: Benjamin Geer <ben {AT} socialtools.net>
Subject: Re: <nettime> floss enforcement/compliance

ed phillips wrote:
 > First of all,
 > releasing a trivial hack is in no way compelled by the GPL. [snip]
> No one is compelled
> to release the specific uses they make of free software tools, unless
> they distribute the specific use to the general public.

I know; I wasn't arguing the contrary.  I was suggesting that if the 
source code of non-trivial programs was published as a matter of course 
(regardless of what the GPL requires), this would in the main be 
beneficial, because it would help reduce the number of hacks and promote 
public ownership of knowledge.

> Yes, I've heard these IBM fellows on numerous panels give the same
> kind of laughter producing schtick. But, what trivial hacks have you
> actually seen IBM release to the networks?

That's precisely my point, and IBM's point.  They seemed to be saying 
that their code quality had increased *because* they were developing 
free software.  IBM has a reputation to protect.

> But please don't just release code you were too lazy or too harried to
> review or to make generic enough to be of use to anyone else.

Fine.  On the other hand, there are large, well-written applications, 
representing an immense amount of knowledge and problem-solving, which 
are never released outside the organisations that wrote them, and whose 
source code the GPL cannot therefore give us access to.  I think it 
would be in the public interest for the source code of those 
applications to be released, too.

> As someone who is out here promoting and using free software in the actually
> existing networks in more than a merely academic way, this aspect of
> the GPL has been one of the most important. The GPL has allowed me to
> tell my clients, whether governmental or private, that their specific
> code is theirs in the sense that they are not compelled to release it
> to the general public but that none of the generic problems that I solve
> belong to them or can be considered their intellectual property, and
> that in fact generic tools and solutions should be contributed back to
> the community of users from which they have benefitted.

You can of course tell them that, but nothing stops them from developing 
high-quality generic tools without your help, and keeping that source 
code to themselves; that's what my employer does.  This, I think, is a 
flaw in the system.  A lot of companies these days are benefitting 
immensely from free software without contributing anything at all to its 
development.  They do this either by using only BSD- or Apache-licensed 
free software, or by using free software as part of programs that they 
don't distribute; companies like Amazon, Yahoo and Google fall into the 
latter category.

It's perhaps easier to see that citizens ought to be able to study 
source code developed by (or for) governments; this follows from the 
principle that we ought to be able to find out what our governments are 
doing, in order to hold them accountable.  (See the recent controversy 
about electronic voting machines for an example of why.)

In the case of software companies, it seems to me that when a program 
embodies a non-trivial amount of knowledge that isn't readily available 
elsewhere, the benefit to society of having access to that knowledge 
outweighs the benefit to a company's shareholders of being able to 
forbid access to it.

> We would never have convinced the large number of private and
> governmental users of free software that it made sense to them to use
> it if they were compelled to release any and all software that they
> develop for their own use.

It is often difficult to persuade people that their personal gain is in 
conflict with the public interest.  Perhaps we just need stronger forms 
of persuasion.  Like laws.

Ben

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

#  distributed via <nettime>: no commercial use without permission
#  <nettime> is a moderated mailing list for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info: majordomo {AT} bbs.thing.net and "info nettime-l" in the msg body
#  archive: http://www.nettime.org contact: nettime {AT} bbs.thing.net